Distance-based reconstruction of tree models for oncogenesis.

Comparative genomic hybridization (CGH) is a laboratory method to measure gains and losses in the copy number of chromosomal regions in tumor cells. It is hypothesized that certain DNA gains and losses are related to cancer progression and that the patterns of these changes are relevant to the clinical consequences of the cancer. It is therefore of interest to develop models which predict the occurrence of these events, as well as techniques for learning such models from CGH data. We continue our study of the mathematical foundations for inferring a model of tumor progression from a CGH data set that we started in Desper et al. (1999). In that paper, we proposed a class of probabilistic tree models and showed that an algorithm based on maximum-weight branching in a graph correctly infers the topology of the tree, under plausible assumptions. In this paper, we extend that work in the direction of the so-called distance-based trees, in which events are leaves of the tree, in the style of models common in phylogenetics. Then we show how to reconstruct the distance-based trees using tree-fitting algorithms developed by researchers in phylogenetics. The main advantages of the distance-based models are that 1) they represent information about co-occurrences of all pairs of events, instead of just some pairs, 2) they allow quantitative predictions about which events occur early in tumor progression, and 3) they bring into play the extensive methodology and software developed in the context of phylogenetics. We illustrate the distance-based tree method and how it complements the branching tree method, with a CGH data set for renal cancer.     

Analysis of wilms tumors using SNP mapping array-based comparative genomic hybridization

Wilms tumor (WT) has been a model to study kidney embryogenesis and tumorigenesis and, although associated with hereditary, cancer predisposition syndromes, the majority of tumors occur sporadically. To analyze genetic changes in WT we have defined copy number changes and loss of heterozygosity in 56 Wilms tumors using high resolution oligonucleotide arrays at a average resolution of ~12 Kb. Consistent deletions were seen on chromosomes 1p, 4q, 7p, 9q, 11p, 11q, 14q, 16q, and 21q. High frequency gains were seen for 1q and lower frequency gains were seen on 7q and chromosomes 8, 12 and 18. The high resolution provided by the SNP mapping arrays has defined minimal regions of deletion for many of these LOH events. Analysis of CNAs by tumor stage show relatively stable karyotypes in stage 1 tumors and more complex aCGH profiles in tumors from stages 3-5.     

Chromosomal imbalances in primary and metastatic melanomas revealed by comparative genomic hybridization

Characteristic genetic changes underlying the metastatic progression of malignant melanoma is incompletely understood. The goal of our study was to explore specific chromosomal alterations associated with the aggressive behavior of this neoplasm. Comparative genomic hybridization was performed to screen and compare genomic imbalances present in primary and metastatic melanomas. Sixteen primary and 12 metastatic specimens were analyzed. We found that the pattern of chromosomal aberrations is similar in the two subgroups; however, alterations present only in primary and/or metastatic tumors were also discovered. The mean number of genetic changes was 6.3 (range 1-14) in primary and 7.8 (range 1-16) in metastatic lesions. Frequent losses involved 9p and 10q, whereas gains most often occurred at 1q, 6p, 7q, and 8q. Distinct, high-level amplifications were mapped to 1p12-p21 and 1p22-p31 in both tumor types. Amplification of 4q12-q13.1, 7q21.3-qter and 8q23-qter were detected only in primary tumors. The 20q13-qter amplicon was present in a metastatic tumor. The number of genetic alterations were significantly higher in primary tumors which developed metastases within one year after the surgery compared to tumors without metastasis during this time period. Fluorescence in situ hybridization with centromeric and locus-specific probes was applied to validate CGH results on a subset of tumors. Comparison of FISH and CGH data gave good correlation. The aggressive behavior of melanoma is associated with accumulation of multiple genetic alterations. Chromosome regions, which differ in the primary and metastatic lesions, may represent potential targets to identify metastases-related chromosomal alterations.     

Clinical and Histopathological Characterization of Cutaneous Melanomas in the Melanoblastoma‐Bearing Libechov Minipig Model

Spontaneous animal tumors appear to be highly suitable models to study human oncology and cancer therapy. The aim of this study was to characterize the clinical and histological features of hereditary melanocytic lesions found in the French herd of melanoblastoma‐bearing Libechov minipigs (MeLiM) and their Duroc crossbreeds. Clinically, we discriminated between three types of melanocytic skin lesions, which offer a lesion continuum from lentigo to metastatic melanomas. More than 70% of these lesions appear on piglets before they are 3 months old and preferentially on homogeneous black coat piglets. The incidence of melanoma reaches 50% in MeLiM. Most of the highly invasive melanomas regressed spontaneously in the first year of the piglet's life and the regression was followed by hair, skin and iris depigmentation. A histopathological study was conducted according to the human melanoma classification. Except for lentigo maligna, we observed the three main types of human melanoma in swine [superficial spreading melanoma (SSM), nodular or unclassified melanoma] with an excess of SSM (59–67%). The histological events leading to total spontaneous regression are chronologically described. The genetic predisposition, the high incidence of melanoma, the clinical and histopathological features similar to the human disease and the high rate of spontaneous regression offer an opportunity to use this model for studying genetic events controlling melanoma development and regression and the biological mechanisms involved in oncogenesis and anti‐cancerous self‐defense.     

Clinical and histopathological characterization of cutaneous melanomas in the melanoblastoma-bearing Libechov minipig model

Spontaneous animal tumors appear to be highly suitable models to study human oncology and cancer therapy. The aim of this study was to characterize the clinical and histological features of hereditary melanocytic lesions found in the French herd of melanoblastoma-bearing Libechov minipigs (MeLiM) and their Duroc crossbreeds. Clinically, we discriminated between three types of melanocytic skin lesions, which offer a lesion continuum from lentigo to metastatic melanomas. More than 70% of these lesions appear on piglets before they are 3 months old and preferentially on homogeneous black coat piglets. The incidence of melanoma reaches 50% in MeLiM. Most of the highly invasive melanomas regressed spontaneously in the first year of the piglet's life and the regression was followed by hair, skin and iris depigmentation. A histopathological study was conducted according to the human melanoma classification. Except for lentigo maligna, we observed the three main types of human melanoma in swine [superficial spreading melanoma (SSM), nodular or unclassified melanoma] with an excess of SSM (59-67%). The histological events leading to total spontaneous regression are chronologically described. The genetic predisposition, the high incidence of melanoma, the clinical and histopathological features similar to the human disease and the high rate of spontaneous regression offer an opportunity to use this model for studying genetic events controlling melanoma development and regression and the biological mechanisms involved in oncogenesis and anti-cancerous self-defense.     

The genetics of uveal melanoma: an emerging framework for targeted therapy

Uveal melanoma is the second most common form of melanoma and the most common primary intraocular malignancy. Until recently, very little was known about the genetics of this aggressive cancer. Mutations in oncogenes and tumor suppressors that are common in other cancers are conspicuously absent in uveal melanoma. In recent years, however, uveal melanoma has begun to yield its secrets, and a fascinating picture is emerging of how it develops and progresses. Mutations in the G(q) alpha subunits, encoded by GNAQ and GNA11, appear to be early or perhaps initiating events that require further mutations for malignant transformation. On the other hand, mutations in the BRCA1-associated protein-1 (BAP1) appear to occur later and demarcate a molecular brink beyond which metastasis becomes highly likely. BAP1 mutations can also occur in the germline, leading to a distinctive cancer predisposition syndrome. These mutations appear to be key events that provide the potential for targeted therapy. This article will review the genetic findings in uveal melanoma over the past two decades and suggest important areas for future work.     

Selective genomic copy number imbalances and probability of recurrence in early-stage breast cancer

A number of studies of copy number imbalances (CNIs) in breast tumors support associations between individual CNIs and patient outcomes. However, no pattern or signature of CNIs has emerged for clinical use. We determined copy number (CN) gains and losses using high-density molecular inversion probe (MIP) arrays for 971 stage I/II breast tumors and applied a boosting strategy to fit hazards models for CN and recurrence, treating chromosomal segments in a dose-specific fashion (-1 [loss], 0 [no change] and +1 [gain]). The concordance index (C-Index) was used to compare prognostic accuracy between a training (n = 728) and test (n = 243) set and across models. Twelve novel prognostic CNIs were identified: losses at 1p12, 12q13.13, 13q12.3, 22q11, and Xp21, and gains at 2p11.1, 3q13.12, 10p11.21, 10q23.1, 11p15, 14q13.2-q13.3, and 17q21.33. In addition, seven CNIs previously implicated as prognostic markers were selected: losses at 8p22 and 16p11.2 and gains at 10p13, 11q13.5, 12p13, 20q13, and Xq28. For all breast cancers combined, the final full model including 19 CNIs, clinical covariates, and tumor marker-approximated subtypes (estrogen receptor [ER], progesterone receptor, ERBB2 amplification, and Ki67) significantly outperformed a model containing only clinical covariates and tumor subtypes (C-Index _{full model}, train[test]  =  0.72[0.71] ± 0.02 vs. C-Index _{clinical + subtype model}, train[test]  =  0.62[0.62] ± 0.02; p<10⁻⁶). In addition, the full model containing 19 CNIs significantly improved prognostication separately for ER–, HER2+, luminal B, and triple negative tumors over clinical variables alone. In summary, we show that a set of 19 CNIs discriminates risk of recurrence among early-stage breast tumors, independent of ER status. Further, our data suggest the presence of specific CNIs that promote and, in some cases, limit tumor spread.     

Multicolor FISH analysis of chromosomal breaks, duplications, deletions, and numerical abnormalities in the sperm of healthy men

Transmitted de novo structural chromosomal abnormalities, the majority of which are paternally derived, can lead to abnormal reproductive outcomes as well as genetic diseases in offspring. We developed and validated a new multicolor FISH procedure (sperm ACM, which utilizes DNA probes specific for the alpha [1cen], classical, [1q12], and midi [1p36.3] satellites of chromosome 1) which utilizes DNA probes specific for three regions of chromosome 1 to detect human sperm that carry numerical abnormalities plus two categories of structural aberrations: (1) duplications and deletions of 1pter and 1cen, and (2) chromosomal breaks within the 1cen-1q12 region. In healthy men, the average frequencies of sperm with duplications and deletions were (a) 4.5 +/- 0.5 and 4.1 +/- 1.3 per 10(4) involving 1pter and (b) 0.9 +/- 0.4 and 0.8 +/- 0.3 per 10(4) involving 1cen, respectively. The frequency of sperm exhibiting breaks within the 1cen-1q12 region was 14.1 +/- 1.2 per 10(4). Structural aberrations accounted for 71% of the abnormalities detected by sperm ACM, which was significantly higher than numerical abnormalities (P=2x10-8). Our findings also suggest that, for healthy men, (a) sperm carrying postmeiotic chromosomal breaks appear to be more prevalent than those carrying products of premeiotic or meiotic breakage or rearrangements, (b) the high frequency of chromosome breaks measured after "fertilization" by the hamster-egg cytogenetic method already appear to be present and detectable within human sperm by FISH, and (c) there are nonrandom and donor-specific distributions of breakpoint locations within 1q12 in sperm. FISH facilitates the analysis of much larger numbers of sperm than was possible when the hamster-egg method was used. Therefore, FISH-based procedures for simultaneously detecting chromosomal breaks, rearrangements, and numerical abnormalities in sperm may have widespread applications in human genetics, genetic toxicology, and reproductive medicine.     

Marked differences in unilateral isolated retinoblastomas from young and older children studied by comparative genomic hybridization

Although it is established that the loss of function of both alleles of the RB1 gene is a prerequisite for the development of retinoblastoma, little is known about the genetic events that are required for tumor progression. We used comparative genomic hybridization (CGH) to search for DNA copy number changes in isolated unilateral retinoblastomas. From a series of 66 patients with retinoblastomas with somatic mutations in both RB1 alleles, tumor samples from 13 children with the youngest (2.0-9.8 months) and 13 with the oldest (36.2-84.1 months) age at operation were studied. Loss at 13q14, the location of RB1, was demonstrated in two tumors only. Recurring chromosome imbalances included gains at 6p (11/26), 1q (10/26), 2p (4/26), and 17q (4/26), gains of the entire chromosome 19 (3/26), and losses at 16q (9/26). A commonly gained region at 1q32 was identified. Increased dosage of GAC1, a candidate oncogene located in 1q32, was found in two of four tumors by Southern blot analysis. Comparison of the CGH findings revealed that retinoblastomas from children with an older age at operation showed significantly more frequent (13/13 cases vs 4/13 cases; P = 0.0005) and more complex genetic abnormalities (median, 5 changes/abnormal tumor vs median, 1.5 changes/abnormal tumor; P = 0.003) than retinoblastomas from children with a young age at operation. Gains at 1q, 2p, 17q, of the entire chromosome 19 and losses of 16q were restricted to the older age group. Our results suggest that the progression of retinoblastomas from older patients follows mutational pathways different from those of younger patients.     

From Mendelian to molecular genetics: the Xiphophorus melanoma model

In human tumor biopsies it is almost impossible to pinpoint the particular molecular abnormalities that determine neoplasia. In animal models where tumorigenesis is initiated by clearly defined genetic events, it is possible to study the genes and their functions that make a normal cell become a fully malignant cancer cell. In the fish Xiphophorus, melanoma can be initiated by simple crossings, and the signaling pathways that govern tumor growth and progression can be delineated. This model offers the prospect of obtaining a complete picture of the molecular changes and regulatory networks underlying tumor formation, which should contribute to a better understanding of some general principles of cancer biology, and identify new targets for melanoma research in particular.     

Mtreemix: a software package for learning and using mixture models of mutagenetic trees

SUMMARY: Mixture models of mutagenetic trees constitute a class of probabilistic models for describing evolutionary processes that are characterized by the accumulation of permanent genetic changes. They have been applied to model the accumulation of chromosomal gains and losses in tumor development and the development of drug resistance-associated mutations in the HIV genome.Mtreemix is a software package for estimating mutagenetic trees mixture models from observed cross-sectional data and for using these models for predictions. We provide programs for model fitting, model selection, simulation, likelihood computation and waiting time estimation. AVAILABILITY: Mtreemix, including source code, documentation, sample data files and precompiled Solaris and Linux binaries, is freely available for non-commercial users at http://mtreemix.bioinf.mpi-sb.mpg.de/     

Molecular cytogenetic analysis of follicular lymphoma (FL) provides detailed characterization of chromosomal instability associated with the t(14;18)(q32;q21) positive and negative subsets and histologic progression

We analyzed a cohort of 61 follicular lymphomas (FL) with an abnormal G-banded karyotype by spectral karyotyping (SKY) to better define the chromosome instability associated with the t(14;18)(q32;q21) positive and negative subsets of FL and histologic grade. In more than 70% of the patients, SKY provided additional cytogenetic information and up to 40% of the structural abnormalities were revised. The six most frequent breakpoints in both SKY and G-banding analyses were 14q32, 18q21, 3q27, 1q11-q21, 6q11-q15 and 1p36 (15-77%). SKY detected nine additional sites (1p11-p13, 2p11-p13, 6q21, 8q24, 6q21, 9p13, 10q22-q24, 12q11-q13 and 17q11-q21) at an incidence of >10%. In addition to the known recurring translocations, t(14;18)(q32;q21) [70%], t(3;14)(q27;q32) [10%], t(1;14)(q21;q32) [5%] and t(8;14)(q24;q32) [2%] and their variants, 125 non-IG gene translocations were identified of which four were recurrent within this series. In contrast to G-banding analysis, SKY revealed a greater degree of karyotypic instability in the t(14;18) (q32;q21) negative subset compared to the t(14;18)(q32;q21) positive subset. Translocations of 3q27 and gains of chromosome 1 were significantly more frequent in the former subset. SKY also allowed a better definition of chromosomal imbalances, thus 37% of the deletions detected by G-banding were shown to be unbalanced translocations leading to gain of genetic material. The majority of recurring (>10%) imbalances were detected at a greater (2-3 fold) incidence by SKY and several regions were narrowed down, notably at gain 2p13-p21, 2q11-q21, 2q31-q37, 12q12-q15, 17q21-q25 and 18q21. Chromosomal abnormalities among the different histologic grades were consistent with an evolution from low to high grade disease and breaks at 6q11-q15 and 8q24 and gain of 7/7q and 8/8q associated significantly with histologic progression. This study also indicates that in addition to gains and losses, non-IG gene translocations involving 1p11-p13, 1p36, 1q11-q21, 8q24, 9p13, and 17q11-q21 play an important role in the histologic progression of FL with t(14;18)(q32;q21) and t(3q27).     

MLH1 and MSH2 promote the symmetry of double-strand break repair events at the HIS4 hotspot in Saccharomyces cerevisiae

Double-strand breaks (DSBs) initiate meiotic recombination. The DSB repair model predicts that both genetic markers spanning the DSB should be included in heteroduplex DNA and be detectable as non-Mendelian segregations (NMS). In experiments testing this, a significant fraction of events do not conform to this prediction, as only one of the markers displays NMS (one-sided events). Two explanations have been proposed to account for the discrepancies between the predictions and experimental observations. One suggests that two-sided events are the norm but are "hidden" as heteroduplex repair frequently restores the parental configuration of one of the markers. Another explanation posits that one-sided events reflect events in which heteroduplex is formed predominantly on only one side of the DSB. In the absence of heteroduplex repair, the first model predicts that two-sided events would be revealed at the expense of one-sided events, while the second predicts no effect on the distribution of events when heteroduplex repair is lost. We tested these predictions by deleting the DNA mismatch repair genes MSH2 or MLH1 and analyzing the proportion of two-sided events. Unexpectedly, the results do not match the predictions of either model. In both mlh1Delta and msh2Delta, the proportion of two-sided events is significantly decreased relative to wild type. These observations can be explained in one of two ways. Either Msh2p/Mlh1p-independent mispair removal leads to restoration of one of the markers flanking the DSB site or Msh2p/Mlh1p actively promote two-sided events.     

Estimating cancer survival and clinical outcome based on genetic tumor progression scores

MOTIVATION: In cancer research, prediction of time to death or relapse is important for a meaningful tumor classification and selecting appropriate therapies. Survival prognosis is typically based on clinical and histological parameters. There is increasing interest in identifying genetic markers that better capture the status of a tumor in order to improve on existing predictions. The accumulation of genetic alterations during tumor progression can be used for the assessment of the genetic status of the tumor. For modeling dependences between the genetic events, evolutionary tree models have been applied. RESULTS: Mixture models of oncogenetic trees provide a probabilistic framework for the estimation of typical pathogenetic routes. From these models we derive a genetic progression score (GPS) that estimates the genetic status of a tumor. GPS is calculated for glioblastoma patients from loss of heterozygosity measurements and for prostate cancer patients from comparative genomic hybridization measurements. Cox proportional hazard models are then fitted to observed survival times of glioblastoma patients and to times until PSA relapse following radical prostatectomy of prostate cancer patients. It turns out that the genetically defined GPS is predictive even after adjustment for classical clinical markers and thus can be considered a medically relevant prognostic factor. AVAILABILITY: Mtreemix, a software package for estimating tree mixture models, is freely available for non-commercial users at http://mtreemix.bioinf.mpi-sb.mpg.de. The raw cancer datasets and R code for the analysis with Cox models are available upon request from the corresponding author.     

Alterations of the biochemical pathways of plants by the air pollutant ozone: which are the true gauges of injury?

Plant strategies to survive ozone stress include exclusion or tolerance of ozone. If these processes fail, past observations of ozone injury have indicated many physiological and metabolic changes then occur; most of these changes are likely to have been initiated at the level of gene expression, suggesting signal transduction. In the last decade considerable understanding of the biochemical process within plants has been developed. Currently there are several hypotheses regarding a response of plants to ozone fumigation: [1] membrane dysfunction and alteration of purpose; [2] stress ethylene interactions; [3] impairment of photosynthesis via changes in Rubisco levels and the guard cells so that the stomata do not track correctly the environment; [4] antioxidant protection through metabolites and enzyme systems to reduce the oxidant load; and [5] general impairment or disruption of metabolic pathways. Many believe that free radicals and other oxidative products, formed in plant leaves under ozone exposure, are responsible for much of the spread of the biochemical alterations. There are obvious chemicals that may account for the changes that are observed, such as hydrogen peroxide. Once the ozone enters the tissue, evidence suggests the first line of defense is a range of antioxidants, such as ascorbate, glutathione peroxidase, superoxide dismutase, and catalase. If overwhelmed, subsequent events occur which are highly suggestive of systemic acquired resistance. Furthermore, other defensive indicators, such as salicylic acid and jasmonic acid, tend to increase, but more slowly than ethylene, and spread their signaling effects more widely in the plant. The primary set of metabolic reactions that ozone triggers is thought to be "wounding" responses with a secondary response of senescence. The dramatic strides in understanding the genetic make-up of plants, gene control, and signal transduction/control over the last few years will only accelerate in the future. We need now to have an understanding of those events that can be translated into more detailed schemes of how ozone alters much of the basic metabolism of plants and how plants counteract or cope with ozone. What is now known about how varied biochemicals and their pathways are changed upon ozone exposure will be discussed.     

Molecular Markers of Tumor Progression in Melanoma

Malignant melanoma represents one of the most aggressive malignancies but outcome is highly variable with early tumor lesions having an excellent prognosis following resection. We review here the data on identification of genes involved in the progression of melanoma as a result of expression array studies, genomic profiling, and genetic models. We focus on the role of tumor suppressors involved in cell cycle function, DNA repair, and genome maintenance. Highlighted are the roles of loss of p16 in promoting neoplasia in cooperation with deregulated MAPK signaling, and the role of loss of the RASSF1A protein in promoting chromosomal instability. The interactions between point mutation in growth signaling molecules and epigenetic changes in genes involved in DNA repair and cell division are discussed.Key Words: Melanoma, genetic progression, tumor suppressors, cell cycle, alkylating agents, mitotic spindle, mitotic spindle poisons.     

Unlike p53, p27 failed to exhibit an anti-tumor genetic interaction with Ku80

Ku80 is often referred to as a tumor suppressor since it maintains the genome by repairing DNA double-strand breaks (DSBs) via the nonhomologous end joining (NHEJ) pathway. Even though Ku80 deletion causes hypersensitivity to γ-radiation, DNA damage and chromosomal rearrangements, Ku80-mutant mice exhibit very low cancer levels. We previously hypothesized these low cancer levels were caused by enhanced cell cycle checkpoints that responded to inefficiently repaired DNA damage because Ku80-mutant fibroblasts exhibit premature cellular senescence that was dependent on a p53-mediated DNA damage response. In addition, Ku80 and p53 show a genetic interaction to suppress pro-B cell lymphoma and medulloblastoma. Here we tested for a similar anti-tumor genetic interaction between Ku80 and the cyclin kinase inhibitor, p27^Kip1 (p27) since p27 mutant mice showed elevated levels of pituitary adenoma that were exacerbated by γ-radiation-induced DNA damage (damage repaired by Ku80). We found that deleting both Ku80 and p27 did not exacerbate cancer as compared to either single mutant. In addition, fibroblasts deleted for both exhibited premature cellular senescence similar to Ku80-mutant fibroblasts. Thus, p27 did not exhibit an obvious genetic interaction with Ku80 to suppress tumors. This observation suggests that DNA damage (or DNA damage responses) induced by either γ-radiation or Ku80 deletion are not equivalent since γ-radiation exacerbates oncogenesis in mice deleted for either p53 or p27 while Ku80 deletion exacerbates oncogenesis for only the former genotype.     

Comparative genomic hybridization study of nasal-type NK/T-cell lymphoma

BACKGROUND: Nasal-type NK/T-cell lymphoma is a rare type of non-Hodgkin's lymphoma. The genetic changes associated with pathogenesis have not been well defined. This study investigates the nonrandom genetic alteration of nasal-type NK/T-cell lymphoma. METHODS: Nine cases were studied. Comparative genomic hybridization (CGH) was carried out using fresh tumor tissues of seven nasal-type NK/T-cell lymphomas. To complement the data by CGH, loss of heterozygosity (LOH) of chromosomes 6q, 1p, and 17p using polymorphic markers and p53 gene mutation by polymerase chain reaction-single-strand conformation polymorphism (PCR-SSCP) were analyzed. RESULTS: The DNA copy number changes of seven nasal-type NK/T-cell lymphomas were gains on chromosomes 2q(5), 13q(4), 10q(3), 21q(2), 3q(2), 5q(2), and 17q(2), and losses involving chromosomes 1p(4), 17p(4), 12q(3), 13q(2), and 6q(1). One of six cases informative for at least two markers for chromosome 6q showed LOH at D6S300, D6S1639, D6S261, D6S407, and D6S292. Two cases showing loss of 1p and 17q by CGH revealed LOH at D1S214, D1S503, and D17S559. P53 mutation was detected in exon 8 in one of nine cases. CONCLUSIONS: Frequent DNA losses at 1p, 17p, and 12q and gains at 2q, 13q, and 10q suggested that these regions could be targets for further molecular genetic analysis to investigate tumor suppressor genes or oncogenes associated with tumorigenesis of NK/T-cell lymphoma. Infrequent alteration of 6q contrary to previous studies raises doubt about an implication of 6q loss in the pathogenesis of early-stage NK/T-cell lymphoma. Further studies on more defined cases are required to verify their association.